Prediction of the land price for housing development in the capital using Deep Learning Technique: Findings from Thailand

Due to high population density and limited numbers of land availability in Bangkok, the capital of Thailand, land values have been increasing every year. For the success of business projects, investors and real estate developers need accurate land price evaluation and inaccurate analysis can lead to losses for real estate developers, project residents, and surrounding communities. However, this process requires extensive knowledge and experience. This research aimed to present an approach for analyzing land values in Bangkok using Deep Learning Techniques, which can help real estate developers assess appropriate land values more accurately and precisely. The study collected vacant land data in Bangkok through a feasibility study online database and analyzed them by Deep Learning Techniques. The results showed 26 determinants from 5 groups. The study conducted 60 parameter adjustments with a ratio of 24:11, using a Quadratic Loss Function. The Deep Learning Model resulted in an R2 value of 0.844 and an RMSE of 132,079.67 Baht ($3,896.16 USD). The results of this research can be used as an effective decision-making tool for real estate developers, as well as landowners and brokers, in determining appropriate buying or selling prices for land. JEL Codes: R31, C45, L85 Keywords: Machine Learning, Deep Learning, Price Forecasting, Land Value, Land Valuation.